Atmos. Chem. Phys., 13, 4529–4541, 2013www.atmos-chem-phys.net/13/4529/2013/doi:10.5194/acp-13-4529-2013© Author(s) 2013. CC Attribution 3.0 License.

EGU Journal Logos (RGB)

Advances in Geosciences

Open A

ccess

Natural Hazards and Earth System

Sciences

Open A

ccess

Annales Geophysicae

Open A

ccessNonlinear Processes

in Geophysics

Open A

ccess

Atmospheric Chemistry

and PhysicsO

pen Access

Atmospheric Chemistry

and Physics

Open A

ccess

Discussions

Atmospheric Measurement

Techniques

Open A

ccess

Atmospheric Measurement

Techniques

Open A

ccess

Discussions

Biogeosciences

Open A

ccess

Open A

ccess

BiogeosciencesDiscussions

Climate of the Past

Open A

ccess

Open A

ccess

Climate of the Past

Discussions

Earth System Dynamics

Open A

ccess

Open A

ccess

Earth System Dynamics

Discussions

GeoscientificInstrumentation

Methods andData Systems

Open A

ccess

GeoscientificInstrumentation

Methods andData Systems

Open A

ccess

Discussions

GeoscientificModel Development

Open A

ccess

Open A

ccess

GeoscientificModel Development

Discussions

Hydrology and Earth System

Sciences

Open A

ccess

Hydrology and Earth System

Sciences

Open A

ccess

Discussions

Ocean Science

Open A

ccess

Open A

ccess

Ocean ScienceDiscussions

Solid Earth

Open A

ccess

Open A

ccess

Solid EarthDiscussions

The Cryosphere

Open A

ccess

Open A

ccess

The CryosphereDiscussions

Natural Hazards and Earth System

Sciences

Open A

ccess

Discussions

ACE-FTS observations of pyrogenic trace species in boreal biomassburning plumes during BORTAS

K. A. Tereszchuk1, G. Gonzalez Abad1,4, C. Clerbaux2,3, J. Hadji-Lazaro3, D. Hurtmans3, P.-F. Coheur3, andP. F. Bernath1,5

1Department of Chemistry, University of York, Heslington, York YO10 5DD, UK2UPMC Univ. Paris 06, Universite Versailles St-Quentin, CNRS/INSU, LATMOS-IPSL, Paris, France3Spectroscopie de l’Atmosphere, Service de Chimie Quantique et de Photophysique, Universite Libre de Bruxelles (U.L.B.),Brussels, Belgium4Harvard-Smithsonian Center for Astrophysics, Atomic and Molecular Physics Division, Cambridge MA 02138, USA5Department of Chemistry & Biochemistry, Old Dominion University, Norfolk VA 23529-0126, USA

Correspondence to:K. A. Tereszchuk (keith.tereszchuk@york.ac.uk)

Received: 26 November 2012 – Published in Atmos. Chem. Phys. Discuss.: 10 December 2012Revised: 21 March 2013 – Accepted: 12 April 2013 – Published: 2 May 2013

Abstract. To further our understanding of the effects ofbiomass burning emissions on atmospheric composition, theBORTAS campaign (BOReal forest fires on Troposphericoxidants over the Atlantic using Aircraft and Satellites) wasconducted on 12 July to 3 August 2011 during the boreal for-est fire season in Canada. The simultaneous aerial, groundand satellite measurement campaign sought to record in-stances of boreal biomass burning to measure the tropo-spheric volume mixing ratios (VMRs) of short- and long-lived trace molecular species from biomass burning emis-sions. The goal was to investigate the connection betweenthe composition and the distribution of these pyrogenic out-flows and their resulting perturbation to atmospheric chem-istry, with particular focus on oxidant species to determinethe overall impact on the oxidizing capacity of the free tro-posphere.

Measurements of pyrogenic trace species in borealbiomass burning plumes were made by the AtmosphericChemistry Experiment Fourier Transform Spectrometer(ACE-FTS) onboard the Canadian Space Agency (CSA)SCISAT-1 satellite during the BORTAS campaign. Eventhough biomass burning emissions are typically confinedto the boundary layer, outflows are often injected into theupper troposphere by isolated convection and fire-relatedconvective processes, thus allowing space-borne instrumentsto measure these pyrogenic outflows. An extensive set of14 molecules – CH3OH, C2H2, C2H6, C3H6O, CO, HCN,

HCOOH, HNO3, H2CO, NO, NO2, OCS, O3, and PAN –have been analysed. Included in this analysis is the calcula-tion of age-dependent sets of enhancement ratios for each ofthe species originating from fires in North America (Canada,Alaska) and Siberia for a period of up to 7 days. Ratio val-ues for the shorter lived primary pyrogenic species decreaseover time primarily due to oxidation by the OH radical as theplume ages and values for longer lived species such as HCNand C2H6 remain relatively unchanged. Increasing negativevalues are observed for the oxidant species, including O3, in-dicating a destruction process in the plume as it ages suchthat concentrations of the oxidant species have dropped be-low their off-plume values.

Results from previous campaigns have indicated that val-ues for the molar ratios of1O3/1CO obtained from themeasurements of the pyrogenic outflow from boreal fires arehighly variable and range from negative to positive, irrespec-tive of plume age. This variability has been attributed topollution effects where the pyrogenic outflows have mixedwith either local urban NOx emissions or pyrogenic emis-sions from the long-range transport of older plumes, thusaffecting the production of O3 within the plumes. The re-sults from this study have identified another potential causeof the variability in O3 concentrations observed in the mea-surements of biomass burning emissions, where evidence ofstratosphere–troposphere exchange due to the pyroconvec-tive updrafts from fires has been identified. Perturbations

Published by Copernicus Publications on behalf of the European Geosciences Union.

4530 K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS

caused by the lofted emissions in these fire-aided convec-tive processes may result in the intrusion of stratospheric airmasses into the free troposphere and subsequent mixing ofstratospheric O3 into the pyrogenic outflows causing fluctua-tions in observed1O3/1CO molar ratios.

1 Introduction

For nearly 40 yr the scientific community has studied theemission of trace constituents from different types of fueland associated atmospheric concentrations, but our knowl-edge remains incomplete, reflecting the heterogeneous andstochastic nature of this process. Biomass burning has an im-portant role in determining the composition of the Earth’ssurface and atmosphere, and in some regions emissions to theatmosphere rival those from fossil fuel combustion. Space-borne observations of land surface and tropospheric compo-sition provided the first glimpse of the large-scale impact ofburning in the global troposphere, where biomass burningevents represent an important source of gases and particlesreleased into the atmosphere (Crutzen et al., 1979; Crutzenand Andreae, 1990; Andreae and Merlet, 2001).

The emissions from biomass burning are recognized tobe a major contributor of chemically active trace gases andaerosols. Wildfires can alter air quality on regional to hemi-spheric scales, and are an important component of the climatesystem. A wide variety of pyrogenic species are emitted, in-cluding carbon dioxide (CO2), carbon monoxide (CO), andhydrogen cyanide (HCN), plus numerous volatile organic hy-drocarbons (VOCs), oxygenated volatile organic compounds(OVOCs) as well as nitrogen, sulfur, and halogen-containingspecies, which are transformed by photochemical processesoccurring during the first few hours in the plume. Thesemolecules significantly alter the distribution of troposphericozone (O3) and influence the oxidizing capacity of the atmo-sphere (Coheur et al., 2007).

The goal of this work is to report and interpret the tropo-spheric mixing ratios of short and long-lived trace molec-ular species from boreal biomass burning emissions ob-tained from infrared solar occultation measurements madewith the Atmospheric Chemistry Experiment Fourier Trans-form Spectrometer (ACE-FTS) instrument on the Cana-dian Space Agency (CSA) SCISAT-1 satellite (Bernath etal., 2005), and serves as an integral part of the BOR-TAS project (BOReal forest fires on Tropospheric ox-idants over the Atlantic using Aircraft and Satellites)(http://www.geos.ed.ac.uk/research/eochem/bortas/).

With the FAAM BAe 146 Atmospheric Research Aircraft(ARA) and ground stations based primarily in eastern andmaritime Canada, one of the primary goals of the BOR-TAS measurement campaign (Palmer et al., 2012) was tosample aged boreal biomass burning plumes from the long-range transport of pyrogenic outflows originating from the

fire activity that is prevalent in Siberia, Alaska, and north-western Canada during the months of July and August. Theseoutflows track from west to east across mainland Canadaand eventually make their way to the North Atlantic. Thedata obtained from BORTAS would serve to complementthe data acquired during the 2008 NASA Arctic Research ofthe Composition of the Troposphere from Aircraft and Satel-lites (ARCTAS) campaign (Hornbrook et al., 2011; Singh etal., 2010), which focused predominantly on measurements ofnascent and young boreal plumes sampled close to the emis-sion source.

Tracer–tracer correlation calculations are carried out be-tween known pyrogenic species to derive sets of age-dependent enhancement ratios (Lefer et al., 1994), i.e. nor-malized excess mixing ratios (Hobbs et al., 2003). Chemicaltransport models (CTMs) can then be used to study the agingand chemical evolution of boreal biomass burning emissionsto quantify the impact on the global troposphere. In addi-tion, the data can be employed to complement the aircraftand ground measurements made during the BORTAS cam-paign. The main benefit of satellite measurements is that theycan provide a global perspective; observations can be madeof the long-range transport of pyrogenic outflows from thesource, where instrument aircraft and ground stations have alimited range.

This work is also part of an in-depth study being con-ducted in an attempt to characterize biomass burning emis-sions remotely by considering criteria such as age and typeof biomass material to further our understanding of the im-pact of biomass burning on atmospheric chemistry. Includedin this study will be the addition of new VOC and OVOCspecies including PAN (Allen et al., 2005) retrieved byACE-FTS to aid in plume differentiation and characteriza-tion of biomass burning emissions in the upper troposphereand lower stratosphere (UTLS).

ACE-FTS is a high-resolution (0.02 cm−1) Fourier trans-form spectrometer used to record limb spectra of the Earth’satmosphere using the solar occultation technique. ACE-FTSuses the Sun as a light source to record transmittance spec-tra during sunrise and sunset occultations when the instru-ment peers through the atmosphere as the satellite rises andsets in its orbit of the Earth. The resulting spectra, with highsignal-to-noise ratios, are recorded through long atmosphericlimb paths (∼ 300 km effective length), thus providing an ad-equate detection threshold for trace species. ACE-FTS hasan excellent vertical resolution of about 2–3 km in the tro-posphere and can measure up to 30 occultations (i.e. sun-rise and sunset events viewed from the orbiting satellite) perday, with each occultation sampling the atmosphere from150 km down to the cloud tops (or 5 km in the absence ofclouds). The locations of ACE-FTS occultations are dic-tated by the 650 km altitude, 74◦ inclination, circular or-bit of SCISAT-1 and the relative position of the Sun. Overthe course of a year, ACE-FTS records atmospheric spec-tra over a large portion of the globe (Bernath et al., 2005).

Atmos. Chem. Phys., 13, 4529–4541, 2013 www.atmos-chem-phys.net/13/4529/2013/

K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS 4531

The instrument has a wide spectral range that covers a re-gion from 750 to 4400 cm−1, allowing for the retrieval ofnumerous molecular species. Through the version 3.0 dataset(http://www.ace.uwaterloo.ca), retrievals are made for over40 species as well as their isotopologues, more than a dozenof which are found in biomass burning emissions. With thelarge suite of molecular species available, and providingnear-global coverage, ACE-FTS is currently the most suit-able remote-sensing instrument for this type of detailed in-vestigation.

A number of studies have been previously conducted usingremote measurements of biomass burning emissions fromACE-FTS (Rinsland et al., 2005, 2007; Coheur et al., 2007)as well as other space-borne instruments (Coheur et al., 2009;Glatthor et al., 2009; Torres et al., 2009; Turquety et al.,2009). Rinsland et al.(2005) first explored the possibility ofusing ACE-FTS to study the emission of longer lived pri-mary pyrogenic species and their long-range transport to es-tablish the effectiveness of ACE-FTS in measuring the emis-sion of trace species from biomass burning. This work wasfollowed up by a more extensive analysis of the elevatedvolume mixing ratios (VMRs) for numerous species emit-ted from young boreal plumes, deriving emission ratios andcalculating tracer-tracer correlation coefficients to demon-strate their enhancement (Rinsland et al., 2007). The re-sults from these preliminary studies precipitated interest inthe retrieval of additional pyrogenic organic species suchas ethene (C2H4), propyne (C3H4), formaldehyde (H2CO),acetone (C3H6O), and peroxyacetylnitrate – abbreviated asPAN (CH3COO2NO2) – and were the first reported detec-tions of these species using infrared solar occultation spec-troscopy from satellites (Coheur et al., 2007). Most recently,ACE-FTS was successfully used to demonstrate its abilityto characterize biomass burning emissions from distinct fueltypes and follow their chemical evolution as the plumes agein the troposphere (Tereszchuk et al., 2011). The results ofthis investigation indicated that space-borne measurementsof biomass burning emissions from different ecosystems canbe differentiated by their unique chemical composition andthat the formation and destruction of pyrogenic trace specieswithin biomass burning plumes can be monitored, thus form-ing the basis for this more extensive work.

2 ACE-FTS detection of boreal biomass burningemissions

2.1 Plume identification

In past studies using ACE-FTS to identify measurements ofbiomass burning plumes, carbon monoxide (CO) was themolecular species employed to find occultations that havesampled pyrogenic emissions through enhanced concentra-tions of CO relative to the background. CO is a well-knownprimary pyrogenic species with an atmospheric lifetime of

K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS 3

tablish the effectiveness of ACE-FTS in measuring the emis-180

sion of trace species from biomass burning. This work wasfollowed up by a more extensive analysis of the elevatedVMRs for numerous species emitted from young borealplumes, deriving emission ratios and calculating tracer-tracercorrelation coefficients to demonstrate their enhancement185

(Rinsland et al., 2007). The results from these prelimi-nary studies precipitated interest in the retrieval of additionalpyrogenic organic species such as ethene (C2H4), propyne(C3H4), formaldehyde (H2CO), acetone (C3H6O) and per-oxyacetylnitrate, abbreviated as PAN (CH3COO2NO2), and190

were the first reported detections of these species using in-frared solar occultation spectroscopy from satellites (Co-heur et al., 2007). Most recently, ACE-FTS was success-fully used to demonstrate its ability to characterize biomassburning emissions from distinct fuel types and follow their195

chemical evolution as the plumes age in the troposphere(Tereszchuk et al., 2011). The results of this investigation in-dicated that space-borne measurements of biomass burningemissions from different ecosystems can be differentiatedbytheir unique chemical composition and that the formation and200

destruction of pyrogenic trace species within biomass burn-ing plumes can be monitored, thus forming the basis for thismore extensive work.

2 ACE-FTS Detection of Boreal Biomass Burning Emis-sions205

2.1 Plume Identification

In past studies using ACE-FTS to identify measurements ofbiomass burning plumes, carbon monoxide (CO) was themolecular species employed to find occultations that havesampled pyrogenic emissions through enhanced concentra-210

tions of CO relative to the background. CO is a well-knownprimary pyrogenic species with an atmospheric lifetime of∼2 months in the free troposphere, but due to it being alsoemitted from numerous anthropogenic sources, another long-lived species that is more specific to biomass burning was215

chosen for plume detection. Hydrogen cyanide (HCN) has alife time of ∼5 months and is emitted almost entirely frombiomass and biofuel combustion. Typical background con-centrations for HCN in the free troposphere have been deter-mined to be 0.225-0.250 ppb (Li et al., 2003; Singh et al.,220

2003). In this work, occultations containing altitude profileswith enhanced tropospheric VMRs of HCN≥0.30 ppbv aredeemed measurements of biomass burning emissions.

2.2 ACE-FTS Data Processing

Version 3.0 of the ACE-FTS dataset, in which the retrieved225

data have been interpolated onto a 1-km altitude grid, wasused for this work. An overview of the retrieval method, errorbudget for retrieved VMRs and the piecewise quadratic inter-polation of the data are outlined by Boone et al. (2005). To

Fig. 1. An example output of the trajectory data obtained usingHYSPLIT. The 3-day back trajectory calculated from occultationss42910 (purple icon), measured on 2 August 2011, is displayedin Google™Earth with the instances of fire activity observedbyMODIS over North America on 1 August 2011. Spatial and tem-poral coincidences of the back trajectory confirm that the emissionsource of the ACE-FTS measurement of enhanced HCN to be theoutflow from the boreal fires in western Ontario. The measurementis of a young plume that is<24 h in age.

identify measurements made in the free troposphere, values230

for heights of the thermal tropopause were obtained from de-rived meteorological products (DMPs) that are produced us-ing the ACE-FTS Version 3.0 dataset (Manney et al., 2007);all measurements that were recorded above the altitude cor-responding to the tropopause in each occultation were iden-235

tified so that they could be excluded during the calculationof the enhancement ratios. The data obtained from belowthe tropopause was then filtered by calculating the medianVMR values for each altitude on the 1-km grid for each ofthe trace species to be studied. Retrievals containing mea-240

surements with molecular VMRs values that were less than10% of the median value at a given altitude were rejected asthis normally indicates retrieval failure. The primary causeof retrieval failure is due to cloud contamination. ACE-FTScan effectively make measurements through thin clouds, but245

measurements of the free troposphere which contain thickclouds in the field of view (FOV) of the instrument can causethe retrieval to fail at these altitudes and the outcome of theretrieval is a deviation of the vertical profile to unrealisticallylow and even negative values for the VMR of a given molec-250

ular species. The median filtering eliminates all measure-ments within each occultation that contain these large devi-ations to assure that the measurements being used are onlythose where the troposphere is cloudless during the time ofmeasurement. The remainder of the data was then filtered255

to remove measurements containing unrealistically large re-trieval errors as they indicate a failed convergence in the re-trieval. To do this, the median absolute deviation (MAD)was calculated from the associated retrieval errors; if there-

Fig. 1. An example output of the trajectory data obtained usingHYSPLIT. The 3-day back trajectory calculated from occultationss42910 (purple icon), measured on 2 August 2011, is displayedin Google™ Earth with the instances of fire activity observed byMODIS over North America on 1 August 2011. Spatial and tempo-ral coincidences of the back trajectory confirm the emission sourceof the ACE-FTS measurement of enhanced HCN to be the outflowfrom the boreal fires in western Ontario. The measurement is of ayoung plume that is< 24 h in age.

∼ 2 months in the free troposphere, but due to it being alsoemitted from numerous anthropogenic sources, another long-lived species that is more specific to biomass burning waschosen for plume detection. Hydrogen cyanide (HCN) has alife time of ∼ 5 months and is emitted almost entirely frombiomass and biofuel combustion. Typical background con-centrations for HCN in the free troposphere have been deter-mined to be 0.225–0.250 ppb (Li et al., 2003; Singh et al.,2003). In this work occultations containing altitude profileswith enhanced tropospheric VMRs of HCN≥ 0.30 ppbv aredeemed measurements of biomass burning emissions.

2.2 ACE-FTS data processing

Version 3.0 of the ACE-FTS dataset, in which the retrieveddata have been interpolated onto a 1 km altitude grid, wasused for this work. An overview of the retrieval method, errorbudget for retrieved VMRs and the piecewise quadratic inter-polation of the data are outlined byBoone et al.(2005). Toidentify measurements made in the free troposphere, valuesfor heights of the thermal tropopause were obtained from de-rived meteorological products (DMPs) that are produced us-ing the ACE-FTS version 3.0 dataset (Manney et al., 2007);all measurements that were recorded above the altitude cor-responding to the tropopause in each occultation were iden-tified so that they could be excluded during the calculationof the enhancement ratios. The data obtained from belowthe tropopause was then filtered by calculating the medianVMR values for each altitude on the 1 km grid for each ofthe trace species to be studied. Retrievals containing mea-surements with molecular VMRs values that were less than

www.atmos-chem-phys.net/13/4529/2013/ Atmos. Chem. Phys., 13, 4529–4541, 2013

4532 K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS

10 % of the median value at a given altitude were rejected asthis normally indicates retrieval failure. The primary causeof retrieval failure is due to cloud contamination. ACE-FTScan effectively make measurements through thin clouds, butmeasurements of the free troposphere which contain thickclouds in the field of view (FOV) of the instrument can causethe retrieval to fail at these altitudes and the outcome of theretrieval is a deviation of the vertical profile to unrealisti-cally low and even negative values for the VMR of a givenmolecular species. The median filtering eliminates all mea-surements within each occultation that contain these large de-viations to assure that the measurements being used are onlythose where the troposphere is cloudless during the time ofmeasurement. The remainder of the data was then filteredto remove measurements containing unrealistically large re-trieval errors as they indicate a failed convergence in theretrieval. To do this the median absolute deviation (MAD)was calculated from the associated retrieval errors; if the re-trieval error at a given altitude was greater than 100 timesthe MAD value calculated from the dataset, they too wererejected. Having filtered the data for any erroneous measure-ments, those occultations which contained HCN altitude pro-files with VMRs of HCN≥ 0.30 ppbv were deemed measure-ments of biomass burning plumes and measurements withVMRs of HCN≤ 0.25 ppbv treated as non-biomass burningor off-plume measurements.

2.3 Determination of emission sources

Once ACE-FTS occultations containing measurements ofbiomass burning emissions were identified, the origins ofplumes measured from instances of boreal forest fires weredetermined. To do this the combined daily fire data recordedby the Moderate-Resolution Imaging Spectroradiometer(MODIS) instruments onboard the Aqua and Terra satelliteswere obtained from the Fire Information for ResourceManagement System (FIRMS) Archive Download Tool(http://earthdata.nasa.gov/data/nrt-data/firms/active-fire-data)available through NASA’s Earth Observing System Dataand Information System (EOSDIS). The MODIS fire countswere used concertedly with back trajectory calculations gen-erated by the Hybrid Single-Particle Lagrangian IntegratedTrajectory (HYSPLIT) (Draxler and Rolph, 2003) modelto locate the sources of boreal biomass burning measuredby ACE-FTS. The altitudes corresponding to the maximumvalue for the VMR of HCN in the ACE-FTS retrieval profileswere used as the starting point in HYSPLIT to calculate backtrajectories of the air mass from the point of measurementto recorded instances of boreal fires. In this way the relativeage of the plume can be determined with an accuracy of±1 day. An example of plume identification determinedfrom the calculated back trajectories made with HYSPLITis seen in Fig.1, where the outflow originating from largeboreal fires burning in western Ontario (2 August 2011)is measured over Lake Huron/Georgian Bay a few hours

4 K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS

trieval error at a given altitude was greater than 100 times260

the MAD value calculated from the dataset, they too wererejected. Having filtered the data for any erroneous measure-ments, those occultations which contained HCN altitude pro-files with VMRs of HCN≥0.30 ppbv were deemed measure-ments of biomass burning plumes and measurements with265

VMRs of HCN≤0.25 ppbv treated as non-biomass burningor off-plume measurements.

2.3 Determination of Emission Sources

Once ACE-FTS occultations containing measurements ofbiomass burning emissions were identified, the origins of270

plumes measured from instances of boreal forest fires weredetermined. To do this, the combined daily fire data recordedby the Moderate-Resolution Imaging Spectroradiometer(MODIS) instruments onboard the Aqua and Terra satelliteswere obtained from the Fire Information for Resource275

Management System (FIRMS) Archive Download Tool(http://earthdata.nasa.gov/data/nrt-data/firms/active-fire-data)available through NASA’s Earth Observing System Dataand Information System (EOSDIS). The MODIS fire countswere used concertedly with back trajectory calculations gen-280

erated by the HYbrid Single-Particle Lagrangian IntegratedTrajectory (HYSPLIT) (Draxler and Rolph, 2003) model tolocate the sources of boreal biomass burning measured byACE-FTS. The altitudes corresponding to the maximumvalue for the VMR of HCN in the ACE-FTS retrieval profiles285

were used as the starting point in HYSPLIT to calculate backtrajectories of the air mass from the point of measurement torecorded instances of boreal fires. In this way, the relativeage of the plume can be determined with an accuracy of±1 day. An example of plume identification determined290

from the calculated back trajectories made with HYSPLITis seen in Fig. 1, where the outflow originating from largeboreal fires burning in western Ontario (2 August 2011)is measured over Lake Huron/Georgian Bay a few hourslater by ACE-FTS sunset occultation ss42910 (measurement295

location represented by the purple icon). It is understood thatby employing this technique to identify plume sources, thereis some uncertainty in the locations of ground-level fires,and thus the associated plume age, due to the fact that firesare often not detected as hotspots and can inject smoke at a300

variety of initial altitudes. The problem of undersamplingwas outlined by Stubenrauch et al. (2010), demonstratingthat the climatological cloud cover over Canada during thesummer months can be quite high, over 50% in westernCanada during the month of July, the height of fire activity305

for that region. The inability of MODIS to penetrate heavycloud cover suggests that many small fires can potentiallygo undetected by active fire detection and sources couldbe erroneously assigned when relying on hotspots alone.To avoid any ambiguity in locations of these sporadic,310

individual hotspots, the ACE-FTS measurements used in thiswork are from the outflows of substantial biomass burning

Fig. 2. IASI CO total column data for the evening orbit over NorthAmerica on 2 August 2011. The figure clearly shows the biomassburning outflow from the fires in western Ontario. The plume, sam-pled during occultation ss42910 over Lake Huron/Georgian Bay(black diamond), is clearly visible and the path of the outflow emu-lates the HYSPLIT back trajectory where the plume follows a south-easterly path from the source over the Great Lakes towards southernOntario. The elevated CO observed over southern Quebec and Mar-itime Canada is coming from a mixture of older plumes originatingfrom the fires in western Ontario and also concurrent fire activitythat was occurring in the Northwest Territories toward the end ofJuly and the beginning of August.

events, such as those observed in western Ontario during theBORTAS campaign; occurrences of continued fire activityin the same geographical region over an extended period315

time (days to weeks), where numerous hotspots repeatedlyappear, on a daily basis, within the same geographical area.Unfortunately, there still exists the caveat that emissionsfrom undetected fires may have a minor contribution tothe plumes being sampled by ACE-FTS and thus introduce320

some uncertainty in the measurement.Google™Earth was used as a visualization tool to bring

ACE-FTS occultations, the MODIS fire data and HYSPLITtrajectories together in order to facilitate the determinationof the sources of biomass burning plumes. Scripts were writ-325

ten in Keyhole Markup Language (KML) in order to importthe locations of ACE-FTS occultations and MODIS fire datainto Google™Earth along with the air mass trajectories fromHYSPLIT, which conveniently provides the option for out-put into the KML format. Visual inspection of the burn re-330

gion was also done to assist in verifying the type of biomassmaterial that is being burned. The near-real time Level-2 imager data from the MODIS Rapid Response System(http://rapidfire.sci.gsfc.nasa.gov/realtime/?calendar), whichcontains the daily MODIS hotspots mapped onto surface im-335

ages of the Earth, was used to identify the burned area toconfirm that only biomass burning from boreal forests is con-tributing to the pyrogenic emissions measured by ACE-FTS.

Since the HYSPLIT model can only serve to give a gen-

Fig. 2. IASI CO total column data for the evening orbit over NorthAmerica on 2 August 2011. The figure clearly shows the biomassburning outflow from the fires in western Ontario. The plume, sam-pled during occultation ss42910 over Lake Huron/Georgian Bay(black diamond), is clearly visible and the path of the outflow emu-lates the HYSPLIT back trajectory where the plume follows a south-easterly path from the source over the Great Lakes towards southernOntario. The elevated CO observed over southern Quebec and Mar-itime Canada is coming from a mixture of older plumes originatingfrom the fires in western Ontario and also concurrent fire activitythat was occurring in the Northwest Territories toward the end ofJuly and the beginning of August.

later by ACE-FTS sunset occultation ss42910 (measurementlocation represented by the purple icon). It is understood thatby employing this technique to identify plume sources, thereis some uncertainty in the locations of ground-level fires,and thus the associated plume age, due to the fact that firesare often not detected as hotspots and can inject smoke ata variety of initial altitudes. The problem of undersamplingwas outlined byStubenrauch et al.(2010), demonstratingthat the climatological cloud cover over Canada during thesummer months can be quite high – over 50 % in westernCanada during the month of July, the height of fire activityfor that region. The inability of MODIS to penetrate heavycloud cover suggests that many small fires can potentiallygo undetected by active fire detection and sources couldbe erroneously assigned when relying on hotspots alone.To avoid any ambiguity in locations of these sporadic,individual hotspots, the ACE-FTS measurements used in thiswork are from the outflows of substantial biomass burningevents, such as those observed in western Ontario during theBORTAS campaign: occurrences of continued fire activityin the same geographical region over an extended periodtime (days to weeks), where numerous hotspots repeatedlyappear, on a daily basis, within the same geographical area.Unfortunately, there still exists the caveat that emissionsfrom undetected fires may have a minor contribution tothe plumes being sampled by ACE-FTS and thus introducesome uncertainty in the measurement.

Atmos. Chem. Phys., 13, 4529–4541, 2013 www.atmos-chem-phys.net/13/4529/2013/

K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS 4533

Google™ Earth was used as a visualization tool to bringACE-FTS occultations, the MODIS fire data and HYSPLITtrajectories together in order to facilitate the determinationof the sources of biomass burning plumes. Scripts were writ-ten in Keyhole Markup Language (KML) in order to importthe locations of ACE-FTS occultations and MODIS fire datainto Google™ Earth along with the air mass trajectories fromHYSPLIT, which conveniently provides the option for out-put into the KML format. Visual inspection of the burn re-gion was also done to assist in verifying the type of biomassmaterial that is being burned. The near-real-time Level-2 imager data from the MODIS Rapid Response System(http://rapidfire.sci.gsfc.nasa.gov/realtime/?calendar), whichcontains the daily MODIS hotspots mapped onto surface im-ages of the Earth, was used to identify the burned area toconfirm that only biomass burning from boreal forests is con-tributing to the pyrogenic emissions measured by ACE-FTS.

Since the HYSPLIT model can only serve to give a gen-eral indication of the direction of the movement of air massesat a given time, to further aid in the verification of the ori-gin of biomass burning plumes, CO total column measure-ments from the Infrared Atmospheric Sounding Interferom-eter (IASI) onboard the MetOp-A satellite (George et al.,2009; Clerbaux et al., 2009) were used to track the out-flow from boreal fires to validate the output calculated fromthe HYSPLIT back trajectories and to ensure that measure-ments made by ACE-FTS are from a single source andnot a mixture of biomass burning outflows from differentsources, which would produce erroneous values in the cal-culations of the age-dependent enhancement ratios. For ex-ample, plumes measured over the North Atlantic can comefrom a number of different sources since it is a region thatcorresponds to the confluence of numerous air masses. Dur-ing the summer in the Northern Hemisphere, fire activityis prevalent in Canada, Alaska, Siberia, and California. Asdemonstrated during the spring (ARCTAS-A) and summer(ARCTAS-B, ARCTAS-CARB) campaigns in 2008 (Horn-brook et al., 2011; Singh et al., 2010), simultaneous outflowsfrom each of these regions will often converge over Canadaand the North Atlantic. The occurrence of any plume mixingfrom different sources would subsequently create uncertaintyin the determination of plume origin and age. Figure.2 showsan example of IASI CO total column measurements madeover Canada on 2 August 2011. In it, a large biomass burn-ing plume can be clearly seen near the Great Lakes originat-ing from the fire activity occurring in western Ontario, whichwas measured by sunset occultation ss42910 (measurementlocation represented by the black diamond). To track the out-flows from multiple sources, animations were created usingmeasurements from the morning and evening passes of IASI.The movements of these outflows were monitored over a pe-riod of days leading up to the time of the ACE-FTS measure-ment to determine the origin of the plumes and whether thereis convergence from more than one source.

The IASI instrument is designed to measure the spectrumemitted by the Earth-atmosphere system in the TIR spectralrange using nadir geometry and provides near-global cover-age twice daily. Measurements are performed from MetOp-Ain a Sun-synchronous polar orbit with a 98.7◦ inclination tothe equator at an altitude of∼ 817 km. The satellite groundtrack is at about 09:30 local time in the morning (and 21:30in the evening). The time to complete an orbit is about 101min, which implies that MetOp makes a little more than 14orbits a day. The IASI instrument observes the Earth up toan angle of 48.3◦ on both sides of the satellite track. Thiscorresponds to 2× 15 mirror positions and a swath of about2× 1100 km. Each instantaneous field of view (3.3◦

× 3.3◦)is composed of 2×2 circular pixels, each corresponding to a12 km diameter footprint on the ground at nadir.

The IASI CO total column data used in this work were ob-tained from the Ether Atmospheric Chemistry Data Centre(http://www.pole-ether.fr). It should be mentioned that IASIalso played an integral role during the BORTAS aircraft cam-paign as near-real-time (< 3 h after measurement) CO totalcolumn and vertical profile data were provided by the LAT-MOS research group (http://iasi-chem.aero.jussieu.fr/) andused for plume tracking and flight planning purposes.

3 Enhancement ratios

Tracer-CO correlation calculations were performed for 14different molecular species to generate sets of age-dependentenhancement ratios. The majority of these species, such asCH3OH, C2H2, C2H6, HCN, HCOOH, HNO3, H2CO, andOCS are known pyrogenic, biomass burning species andhave been previously studied with ACE-FTS. Data prod-ucts for two new, reactive, pyrogenic species have recentlybeen added to the version 3.0 dataset. These are peroxyacetylnitrate (CH3CO·O2NO2, abbreviated as PAN) and acetone(C3H6O). In addition to these species, NO, NO2 and O3 arealso investigated to observe the effect of biomass burningemission on tropospheric oxidants. All occultations used forthe calculations were recorded during the summer months ofthe Northern Hemisphere (JJA) in the 4 yr (2008–2011) lead-ing up to and including the BORTAS campaign, where IASICO data are available to confirm that the ACE-FTS measure-ments made are from a singular source and not from a mix-ture of multiple pyrogenic outflows.

Since the slope of the tracer-CO calculations are obtainedusing the measurements of trace species that are associatedwith the measurements of enhanced HCN (≥ 0.30 ppbv) inthe biomass burning plumes, the slope value thus representsthe enhancement ratio (Lefer et al., 1994) between the twopyrogenic species. The enhancement ratios can be used todetermine trends in the plume photochemistry and can serveas constraints for CTMs to study the aging of biomass burn-ing emissions in the free troposphere. Similar to the emis-sion ratio (Andreae and Merlet, 2001), enhancement ratios

www.atmos-chem-phys.net/13/4529/2013/ Atmos. Chem. Phys., 13, 4529–4541, 2013

4534 K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS

Table 1.Age-dependent enhancement ratios for numerous pyrogenic species determined from the slope of the linear regression of tracer-COcalculations made from measurements of boreal biomass burning plumes originating from both North America (Canada, Alaska) and Siberiafor a period of 7 days. The number of occultations (n) used to calculate the enhancement ratio for each day is given in parenthesis. Thereported error represents the 1-sigma uncertainty of the slope in the linear regression calculations.

Boreal Enhancement Ratios (ppbv ppbv−1)

Compound 0–24 h 24–48 h 48–72 h 72–96 h 96–120 h 120–144 h 144–168 h

North America (n= 15) (n= 14) (n= 10) (n= 16) (n= 8) (n= 7) (n= 6)

CH3OH (2.32± 0.59)× 10−2 (2.39± 0.68)× 10−2 (2.21± 0.46)× 10−2 (1.96± 0.35)× 10−2 (1.95± 0.26)× 10−2 (1.99± 0.40)× 10−2 (1.81± 0.44)× 10−2

C2H2 (3.32± 0.33)× 10−3 (2.53± 0.44)× 10−3 (1.56± 0.18)× 10−3 (1.10± 0.12)× 10−3 (0.98± 0.19)× 10−3 (0.88± 0.15)× 10−3 (0.68± 0.61)× 10−3

C2H6 (6.73± 1.37)× 10−3 (6.80± 1.08)× 10−3 (7.21± 0.80)× 10−3 (7.05± 1.04)× 10−3 (6.92± 0.96)× 10−3 (6.94± 0.84)× 10−3 (6.91± 1.13)× 10−3

C3H6O (1.68± 0.25)× 10−2 (1.25± 0.99)× 10−2 (1.20± 0.42)× 10−2 (0.94± 0.19)× 10−2 (0.88± 0.27)× 10−2 (0.86± 0.18)× 10−2 (0.57± 0.76)× 10−2

HCN (2.01± 0.86)× 10−3 (2.36± 0.79)× 10−3 (2.29± 0.40)× 10−3 (2.19± 0.44)× 10−3 (1.84± 0.43)× 10−3 (2.26± 0.88)× 10−3 (2.49± 1.18)× 10−3

HCOOH (4.91± 0.96)× 10−3 (4.25± 0.82)× 10−3 (3.79± 0.66)× 10−3 (3.34± 0.42)× 10−3 (1.40± 0.44)× 10−3 (1.46± 0.49)× 10−3 (1.25± 1.15)× 10−3

HNO3 (–0.34± 0.16)× 10−2 (–0.45± 0.42)× 10−2 (–0.54± 0.15)× 10−2 (–0.56± 0.13)× 10−2 (–0.65± 0.22)× 10−2 (–0.74± 0.32)× 10−2 (–1.29± 0.19)× 10−2

H2CO (1.33± 0.28)× 10−3 (1.14± 0.33)× 10−3 (0.85± 0.29)× 10−3 (0.72± 0.16)× 10−3 (0.36± 0.22)× 10−3 (0.22± 0.14)× 10−3 (0.21± 0.50)× 10−3

NO (3.85± 1.19)× 10−3 (2.72± 0.45)× 10−3 (1.94± 0.72)× 10−3 (1.70± 1.38)× 10−3 (0.71± 0.98)× 10−3 (0.62± 0.17)× 10−3 (0.80± 0.41)× 10−3

NO2 (–0.24± 0.66)× 10−3 (–0.51± 0.36)× 10−3 (–1.58± 0.69)× 10−3 (–1.69± 0.29)× 10−3 (–1.65± 0.21)× 10−3 (–1.34± 0.80)× 10−3 (–1.75± 0.28)× 10−3

OCS (1.28± 0.29)× 10−3 (1.21± 0.33)× 10−3 (1.21± 0.19)× 10−3 (0.92± 0.22)× 10−3 (0.79± 0.46)× 10−3 (0.30± 0.16)× 10−3 (0.14± 0.35)× 10−3

O3 –0.68± 0.24 –1.17± 0.53 –1.21± 0.24 –1.51± 0.41 –1.65± 0.76 –2.32± 0.43 –2.73± 0.49PAN (5.53± 0.69)× 10−3 (4.69± 0.93)× 10−3 (4.33± 0.63)× 10−3 (4.49± 0.97)× 10−3 (3.45± 0.12)× 10−3 (3.61± 0.54)× 10−3 (3.45± 0.16)× 10−3

Siberia (n= 20) (n= 11) (n= 9) (n= 8) (n= 8) (n= 9) (n= 3)

CH3OH (4.28± 0.43)× 10−2 (3.91± 0.37)× 10−2 (3.52± 0.44)× 10−2 (3.16± 0.58)× 10−2 (3.24± 0.55)× 10−2 (3.18± 0.84)× 10−2 (2.13± 1.10)× 10−2

C2H2 (2.15± 0.24)× 10−3 (1.89± 0.18)× 10−3 (1.55± 0.26)× 10−3 (1.14± 0.21)× 10−3 (1.08± 0.22)× 10−3 (1.06± 0.21)× 10−3 (0.81± 0.43)× 10−3

C2H6 (6.09± 1.01)× 10−3 (5.88± 0.91)× 10−3 (6.07± 0.73)× 10−3 (6.29± 1.84)× 10−3 (6.30± 0.89)× 10−3 (6.12± 1.44)× 10−3 (6.94± 2.34)× 10−3

C3H6O (1.43± 0.36)× 10−2 (1.36± 0.34)× 10−2 (1.24± 0.28)× 10−2 (0.95± 0.54)× 10−2 (0.82± 0.36)× 10−2 (0.69± 0.90)× 10−2 (0.62± 0.55x10−2

HCN (2.91± 0.43)× 10−3 (2.62± 0.22)× 10−3 (2.79± 0.60)× 10−3 (2.96± 0.42)× 10−3 (2.77± 0.93)× 10−3 (2.96± 1.08)× 10−3 (2.70± 1.79)× 10−3

HCOOH (6.41± 0.75)× 10−3 (4.15± 0.47)× 10−3 (3.27± 0.74)× 10−3 (2.33± 0.45)× 10−3 (1.45± 1.02)× 10−3 (1.34)± 1.27)× 10−3 (0.77± 1.81)× 10−3

HNO3 (–0.37± 0.12)× 10−2 (–0.39± 0.19)× 10−2 (–0.44± 0.53)× 10−2 (–0.50± 0.22)× 10−2 (–0.57± 0.30)× 10−2 (–0.63± 0.31)× 10−2 (–1.30± 0.08)× 10−2

H2CO (2.54± 0.28)× 10−3 (1.07± 0.16)× 10−3 (1.22± 0.59)× 10−3 (0.92± 0.23)× 10−3 (0.33± 0.18)× 10−3 (0.27± 0.20)× 10−3 (0.18± 0.52)× 10−3

NO (3.19± 0.56)× 10−3 (1.72± 0.36)× 10−3 (2.77± 0.95)× 10−3 (1.12± 0.58)× 10−3 (1.17± 0.50)× 10−3 (0.57± 0.09)× 10−3 (0.55± 0.86)× 10−3

NO2 (–0.83± 1.56)× 10−3 (–0.87± 1.24)× 10−3 (–0.48± 0.23)× 10−3 (–2.04± 1.39)× 10−3 (–1.92± 2.01)× 10−3 (–2.14± 0.58)× 10−3 (–3.65± 0.70)× 10−3

OCS (7.41± 2.01)× 10−4 (8.46± 1.07)× 10−4 (5.94± 2.44)× 10−4 (5.92± 2.16)× 10−4 (5.58± 2.32)× 10−4 (1.77± 1.87)× 10−4 (0.68± 1.45)× 10−4

O3 –0.79± 0.26 –0.82± 0.24 –1.24± 0.52 –1.58± 0.31 –1.56± 0.53 –2.10± 0.72 –2.01± 0.96PAN (2.63± 0.67)× 10−3 (4.42± 0.89)× 10−3 (3.78± 0.64)× 10−3 (3.34± 0.72)× 10−3 (3.80± 0.94)× 10−3 (2.82± 1.71)× 10−3 (2.69± 2.21)× 10−3

are also calculated by dividing the excess trace species con-centrations measured in a fire plume by the excess con-centration of a simultaneously measured reference gas; COis used as the reference species in this study. To obtainthese excess concentrations, the ambient background con-centrations must be subtracted from the values measured inthe plume, i.e.1X / 1CO where1X indicates the value ofXplume−Xbackground. Alternatively, the enhancement ratio canbe determined as the linear regression slope of the speciesconcentration versus the reference species, which is how theenhancement ratios were obtained in this work. The onlydifference between an emission ratio and enhancement ra-tio is that emission ratios are calculated from measurementsat the time of emission, i.e. plumes with an age of zero.The enhancement ratio is the normalized excess mixing ratio(Hobbs et al., 2003) obtained from plumes which are non-nascent and have aged for a period of time.

Sets of age-dependent enhancement ratios (relative to CO)for a number of trace species measured in boreal biomassburning plumes originating from North America (Canadaand Alaska) and Siberia are listed in Table1. The enhance-ment ratios are calculated using the combined data from indi-vidual measurements of plumes having the same relative age.On occasion, ACE-FTS will sample biomass burning plumesby making sequential measurements along the same pyro-

genic outflow, as demonstrated inTereszchuk et al.(2011);special consideration was not given to such coincidencesand all occultation were treated as separate individual mea-surements in the calculation of the sets of enhancement ra-tios. The ages for the pyrogenic emission range from youngplumes (< 24 h old) to those which have aged for a periodof up to 7 days. Measurements of older plumes become moredifficult to obtain as the further away the outflow travels fromthe source, the more difficult it becomes to identify the originwith certainty, and there is increased likelihood that the out-flow will have mixed with plumes being emitted from otherfires that may be burning concurrently in other regions of theNorthern Hemisphere, thus potentially introducing erroneousmeasurements in the calculation of the enhancement ratios.

In Table1 the calculated molar ratios for HCN from theoccultations containing vertical profiles of elevated HCN(≥ 0.30 ppbv) are reported. Enhancements are observed inHCN and also for each of the other simultaneously retrievedpyrogenic species and are consistent with the elevated HCNVMRs values. As would be expected, the values for theshorter lived primary pyrogenic species decrease over timeas these species are oxidized, mainly by reaction with OH,as the plume ages, where values for longer lived species suchas HCN and C2H6 remain relatively unchanged. Increasingnegative values are observed for the oxidant species, which

Atmos. Chem. Phys., 13, 4529–4541, 2013 www.atmos-chem-phys.net/13/4529/2013/

K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS 4535

include O3, NO2 and HNO3, indicating that these species areundergoing a destruction process in the plume as it ages suchthat concentrations of the reactive species have dropped be-low their off-plume values. The enhancement ratios reportedin Table1 are similar to the results that were observed for bo-real biomass burning emissions in a preliminary analysis thatwas done in the characterization of boreal biomass burningplumes using ACE-FTS byTereszchuk et al.(2011), wherea validation of the enhancement ratios was conducted by as-suming that the enhancement ratios for the long-lived pyro-genic species can be utilized as the effective emission ratiosfor these species in the computation of their emission fac-tors. These emission factors were then compared to the datapublished in the work byAkagi et al.(2011), which containscomprehensive sets of emission factors for biomass burningfrom numerous ecosystems and for domestic biomass burn-ing. Included in the data set are the results obtained from themeasurement of fresh Canadian smoke plumes recorded dur-ing ARCTAS-B (Simpson et al., 2011) and numerous othercampaigns which sought to measure the emissions from bo-real biomass burning. The emission factors obtained from theACE-FTS measurements showed good agreement with theairborne emission factors calculated byAkagi et al.(2011),and were within the reported 1-sigma uncertainty.

As the biomass burning plumes age, the emitted gaseouspyrogenic species within it react and chemical transforma-tions of these primary pyrogenic species take place leadingto the formation of secondary species. Of these secondaryspecies formed in aged plumes, O3 and HNO3 are partic-ularly important. HNO3 is mainly formed by OH-inducedconversion of NO2, which results from rapid conversion ofprimary pyrogenic NO. NO2 is also an important precursor toPAN which forms within a few hours. At low temperatures inthe free troposphere, and in the presence of co-emitted NH3,HNO3 will quickly convert to ammonium nitrate (NH4NO3)(Yokelson et al., 2009; Alvarado et al., 2010). At night, NO2+ O3 can make NO3 and NO2 + NO3 can make N2O5, whichcan hydrolyse to make HNO3. If PAN or alkyl nitrates formin the plume, warming of the plume can cause decomposi-tion and release some NO2 in the aged plume as a secondaryproduct. Organics emitted from biomass burning lead to theformation of secondary O3, an important atmospheric oxi-dant and a precursor of OH radicals (Fiedler et al., 2009; Ak-agi et al., 2011). It is well known that reactions with the hy-droxyl radical (OH), the dominant oxidizing chemical in theatmosphere, drive atmospheric oxidation through reactionswith species emitted from the Earth’s surface. Many of thesemolecules lead to the chemical production of troposphericO3 and other reactive trace gases (Rinsland et al., 2005).

4 O3 production in boreal biomass burning

That being said there is currently a great deal of debate sur-rounding the influence of forest fires on O3 production and

K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS 7

burning plumes using ACE-FTS by Tereszchuk et al. (2011),where a validation of the enhancement ratios was conductedby assuming that the enhancement ratios for the long-livedpyrogenic species can be utilized as the effective emission490

ratios for these species in the computation of their emis-sion factors. These emission factors were then comparedto the data published in the work by Akagi et al. (2011),which contains comprehensive sets of emission factors forbiomass burning from numerous ecosystems and for domes-495

tic biomass burning. Included in the data set, are the resultsobtained from the measurement of fresh Canadian smokeplumes recorded during ARCTAS-B (Simpson et al., 2011)and numerous other campaigns which sought to measurethe emissions from boreal biomass burning. The emission500

factors obtained from the ACE-FTS measurements showedgood agreement with the airborne emission factors calculatedby Akagi et al. (2011), and were within the reported 1-sigmauncertainty.

As the biomass burning plumes age, the emitted gaseous505

pyrogenic species within it react and chemical transforma-tions of these primary pyrogenic species take place leadingto the formation of secondary species. Of these secondaryspecies formed in aged plumes, O3 and HNO3 are partic-ularly important. HNO3 is mainly formed by OH-induced510

conversion of NO2, which results from rapid conversion ofprimary pyrogenic NO. NO2 is also an important precursor toPAN which forms within a few hours. At low temperatures inthe free troposphere, and in the presence of co-emitted NH3,HNO3 will quickly convert to ammonium nitrate (NH4NO3)515

(Yokelson et al., 2009; Alvarado et al., 2010). At night, NO2

+ O3 can make NO3 and NO2 + NO3 can make N2O5, whichcan hydrolyze to make HNO3. If PAN or alkyl-nitrates formin the plume, warming of the plume can cause decomposi-tion and release some NO2 in the aged plume as a secondary520

product. Organics emitted from biomass burning lead to theformation of secondary O3, an important atmospheric oxi-dant, and a precursor of OH radicals (Fiedler et al., 2009;Akagi et al., 2011). It is well known that reactions with thehydroxyl radical (OH), the dominant oxidizing chemical in525

the atmosphere, drive atmospheric oxidation through reac-tions with species emitted from the Earth’s surface. Manyof these molecules lead to the chemical production of tro-pospheric O3 and other reactive trace gases (Rinsland et al.,2005).530

4 O3 Production in Boreal Biomass Burning

That being said, there is currently a great deal of debate sur-rounding the influence of forest fires on O3 production andwhether it is generated in boreal biomass burning plumes.The increasingly negative enhancement ratios for O3 in Ta-535

ble 1 indicate that it undergoes a destruction process withinthe plume in the first few days, the net decrease in concen-tration is most likely dominated by oxidation reactions with

Fig. 3. ACE-FTS VMR profiles for HCN and O3 recorded from ayoung boreal plume (ss42910). Enhancements in the HCN VMRsoccur below 10.5 km as the instrument records through the topofthe pyrogenic outflow. Within the plume, O3 VMRs drop to∼25%of their off-plume values, indicating a destruction process is occur-ring. The high negative correlation (-0.932) and negative enhance-ment ratio (-0.70±0.27) for O3 confirm this observation.

numerous short-lived primary pyrogenic species within theyoung plumes. It was also noted by Alvarado et al. (2010)540

that there was very little clear evidence for O3 formation inthe young boreal plumes measured during ARCTAS-B in ei-ther aircraft or satellite observations and suggested thatthiswas due to 40% of the NOx from the fires being converted toPAN within a few hours of emission thus limiting the imme-545

diate reactivity of the plume.An in-depth review of the production of O3 in wildfires

was recently published by Jaffe and Wigder (2012). In it, asummary of values for∆O3/∆CO obtained from numerouscampaigns are tabulated in order of plume age. Values ob-550

tained from the measurements of the pyrogenic outflow fromboreal fires are highly variable and range from negative topositive, irrespective of plume age. According to Jaffe andWigder (2012), this variability is due to the interplay of nu-merous factors including fire emissions, efficiency of com-555

bustion, chemical and photochemical reactions, aerosol ef-fects on chemistry and radiation, and local and downwindmeteorological patterns. One important factor, in particular,is due to pollution interferences. Studies have demonstratedthat there are O3 enhancements in wildfire plumes that have560

mixed with urban NOx emissions (Jaffe and Wigder, 2012;Singh et al., 2010). In a study of Arctic wildfire plumes,Singh et al. (2010) found molar ratios of∆O3/∆CO of ap-proximately 0.03 and 0.11 for fresh biomass burning plumesand aged plumes mixed with urban emissions, respectively.565

It is not clear whether this enhancement is due to mixingwith urban air masses or also due to a suppression of O3

Fig. 3. ACE-FTS VMR profiles for HCN and O3 recorded from ayoung boreal plume (ss42910). Enhancements in the HCN VMRsoccur below 10.5 km as the instrument records through the top of thepyrogenic outflow. Within the plume, O3 VMRs drop to∼ 25 % oftheir off-plume values, indicating a destruction process is occurring.The high negative correlation (−0.932) and negative enhancementratio (−0.70± 0.27) for O3 confirm this observation.

whether it is generated in boreal biomass burning plumes.The increasingly negative enhancement ratios for O3 in Ta-ble 1 indicate that it undergoes a destruction process withinthe plume in the first few days, the net decrease in concen-tration is most likely dominated by oxidation reactions withnumerous short-lived primary pyrogenic species within theyoung plumes. It was also noted byAlvarado et al.(2010)that there was very little clear evidence for O3 formation inthe young boreal plumes measured during ARCTAS-B in ei-ther aircraft or satellite observations and suggested that thiswas due to 40 % of the NOx from the fires being convertedto PAN within a few hours of emission, thus limiting the im-mediate reactivity of the plume.

An in-depth review of the production of O3 in wildfireswas recently published byJaffe and Wigder(2012). In ita summary of values for1O3/1CO obtained from numer-ous campaigns are tabulated in order of plume age. Val-ues obtained from the measurements of the pyrogenic out-flow from boreal fires are highly variable and range fromnegative to positive, irrespective of plume age. Accordingto Jaffe and Wigder(2012), this variability is due to theinterplay of numerous factors including fire emissions, ef-ficiency of combustion, chemical and photochemical reac-tions, aerosol effects on chemistry and radiation, and localand downwind meteorological patterns. One important fac-tor, in particular, is due to pollution interferences. Studieshave demonstrated that there are O3 enhancements in wild-fire plumes that have mixed with urban NOx emissions (Jaffeand Wigder, 2012; Singh et al., 2010). In a study of Arc-tic wildfire plumes,Singh et al.(2010) found molar ratios of

www.atmos-chem-phys.net/13/4529/2013/ Atmos. Chem. Phys., 13, 4529–4541, 2013

4536 K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS8 K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS

Fig. 4. Monthly median VMR values for O3 in the Northern Hemi-sphere from all filtered ACE-FTS measurements made in the tropo-sphere during a four-year period (2008-2011), without the thresholdon HCN applied. The error bars represent the standard deviation forthe set of measurements used in the calculation. The number of oc-cultations, n, used to calculate each monthly median is reported.The altitude range for the measurements are from 5.5-18.5 km. Ele-vated O3 VMRs in the spring months correspond to the agriculturalpractices in Russia and East Asia. The enhanced O3 in the summermonths is coincident with boreal wildfire activity.

production near the fire source. The sets of age-dependentenhancement ratios calculated in this work were done usingonly measurements of boreal biomass burning plumes which570

were identified as coming from a single source. In addition,the measurements made by ACE-FTS are from plumes whichhave been transported high into the upper troposphere fromisolated convection events and/or fire-aided pyroconvectiveupdrafts from large forest fires. This reduces the potentialin-575

teraction with anthropogenic pollution sources due the likeli-hood of confinement within the boundary layer as is the casewith the emissions from smaller fires.

Plotted in Fig. 3 are the ACE-FTS VMR profiles for HCNand O3 from the young boreal plume recorded in sunset oc-580

cultation ss42910. From the HCN profile, we observe thata rapid enhancement in the HCN concentration occurs be-low 10.5 km as the instrument begins to record through thepyrogenic outflow. From within the plume, a markedly de-creased value in the O3 VMRs (∼25%) is observed when585

compared to the off-plume concentrations recorded from thetropopause at 14.6 km (value obtained from the DMP forss42910) down to 10.5 km. The profiles demonstrate that O3

in young plumes is undergoing a destruction process whileO3 outside of the plume is elevated to values much higher590

than background levels sampled from clean marine air in theupper troposphere over the North Atlantic during months ofminimal fire activity in the Northern Hemisphere (∼55 ppb,

(Jaegle et al., 1999)). Both species correlate highly withCO, demonstrating that they are pyrogenic in nature. The595

high negative correlation for O3 indicates that a destructionprocess is occurring in the young plume. This is corrob-orated by the enhancement ratios with respect to CO cal-culated for both HCN and O3, which are (4.26±0.66)x10-3

and -0.70±0.27, respectively. It should be noted that due600

to the small number of data points available from an indi-vidual occultation, the enhancement ratios reported in Fig. 3were not calculated using the same restrictions used for thesets of age-dependent enhancement ratios in Table 1, whereonly data corresponding to measurements≥0.30 ppbv were605

used. The enhancement ratios in Fig. 3 were obtained usingthe data recorded from 10.5-7.5 km, which correspond to themeasurements from within the plume even though the HCNVMRs for the data recorded at 10.5-9.5 km are below 0.30ppbv.610

The profiles in Fig. 3 and the results for O3 in Table 1are reminiscent of similar findings made by Tereszchuk et al.(2011), where a continuous plume outflow originating fromthe tropical forests in Amazonia on 29 October 2004 wasmeasured by ACE-FTS. The off-plume VMRs for O3 were615

much higher than those recorded in the plume, but remainedlower than the off-plume values recorded nearby at similarlatitudes during the same corresponding time period. In ad-dition, these off-plume VMRs were elevated high above thetypical background values for O3. Both scenarios would indi-620

cate that biomass burning is producing O3, but as a secondarypyrogenic species formed in aged plumes. It was noted byDupont et al. (2012) that the young plumes from Asian wild-fires measured during the ARCTAS-A campaign containedthe precursors to react in the troposphere and produce O3,625

but because plume sampling occurred just after emission,sufficient time was not allowed for O3 production to occur.These observations for O3 production were confirmed dur-ing BORTAS from the analysis of ozone photochemistry ob-served by the FAAM BAe-146 research aircraft (Parrington630

et al., 2013).Since the only other major source of tropospheric O3 is

from anthropogenic pollution (smog), which is primarily re-stricted to the lower troposphere (ground level, boundary la-yar), a simplistic way to demonstrate that boreal biomass635

burning plumes are producing O3 is to analyze the annualtrend for O3 in the middle to upper troposphere in the North-ern Hemisphere and observe whether increases in O3 VMRvalues coincide with the lofted emissions originating fromfire activity. As indicated in the introduction of this work,640

ACE-FTS measurements have a bias such that occultationsonly measure down to 5 km above the Earth’s surface, thuslimiting ACE-FTS measurements to the middle to upper tro-posphere. Fig. 4 plots the monthly median VMRs for O3

in the Northern Hemisphere from all filtered ACE-FTS mea-645

surements, without the threshold on HCN applied, made inthe troposphere during a four year period (2008-2011). Theerror bars represent the calculated 1-sigma standard devia-

Fig. 4. Monthly median VMR values for O3 in the Northern Hemi-sphere from all filtered ACE-FTS measurements made in the tro-posphere during a 4 yr period (2008–2011), without the thresholdon HCN applied. The error bars represent the standard deviationfor the set of measurements used in the calculation. The number ofoccultations,n, used to calculate each monthly median is reported.The altitude range for the measurements are from 5.5–18.5 km. Ele-vated O3 VMRs in the spring months correspond to the agriculturalpractices in Russia and East Asia. The enhanced O3 in the summermonths is coincident with boreal wildfire activity.

1O3/1CO of approximately 0.03 and 0.11 for fresh biomassburning plumes and aged plumes mixed with urban emis-sions, respectively. It is not clear whether this enhancementis due to mixing with urban air masses or also due to a sup-pression of O3 production near the fire source. The sets ofage-dependent enhancement ratios calculated in this workwere done using only measurements of boreal biomass burn-ing plumes which were identified as coming from a singlesource. In addition, the measurements made by ACE-FTS arefrom plumes which have been transported high into the up-per troposphere from isolated convection events and/or fire-aided pyroconvective updrafts from large forest fires. Thisreduces the potential interaction with anthropogenic pollu-tion sources due to the likelihood of confinement within theboundary layer, as is the case with the emissions from smallerfires.

Plotted in Fig.3 are the ACE-FTS VMR profiles for HCNand O3 from the young boreal plume recorded in sunset oc-cultation ss42910. From the HCN profile, we observe thata rapid enhancement in the HCN concentration occurs be-low 10.5 km as the instrument begins to record through thepyrogenic outflow. From within the plume, a markedly de-creased value in the O3 VMRs (∼ 25 %) is observed whencompared to the off-plume concentrations recorded from thetropopause at 14.6 km (value obtained from the DMP forss42910) down to 10.5 km. The profiles demonstrate that O3in young plumes is undergoing a destruction process, while

O3 outside of the plume is elevated to values much higherthan background levels sampled from clean marine air in theupper troposphere over the North Atlantic during months ofminimal fire activity in the Northern Hemisphere (∼ 55 ppb,Jaegle et al., 1999). Both species correlate highly with CO,demonstrating that they are pyrogenic in nature. The highnegative correlation for O3 indicates that a destruction pro-cess is occurring in the young plume. This is corroboratedby the enhancement ratios with respect to CO calculatedfor both HCN and O3, which are (4.26± 0.66)× 10−3 and−0.70± 0.27, respectively. It should be noted that due tothe small number of data points available from an individ-ual occultation, the enhancement ratios reported in Fig.3were not calculated using the same restrictions used for thesets of age-dependent enhancement ratios in Table1, whereonly data corresponding to measurements≥ 0.30 ppbv wereused. The enhancement ratios in Fig.3 were obtained us-ing the data recorded from 10.5–7.5 km, which correspondto the measurements from within the plume even though theHCN VMRs for the data recorded at 10.5–9.5 km are below0.30 ppbv.

The profiles in Fig.3 and the results for O3 in Table 1are reminiscent of similar findings made byTereszchuk et al.(2011), where a continuous plume outflow originating fromthe tropical forests in Amazonia on 29 October 2004 wasmeasured by ACE-FTS. The off-plume VMRs for O3 weremuch higher than those recorded in the plume, but remainedlower than the off-plume values recorded nearby at similarlatitudes during the same corresponding time period. In ad-dition, these off-plume VMRs were elevated high above thetypical background values for O3. Both scenarios would indi-cate that biomass burning is producing O3, but as a secondarypyrogenic species formed in aged plumes. It was noted byDupont et al.(2012) that the young plumes from Asian wild-fires measured during the ARCTAS-A campaign containedthe precursors to react in the troposphere and produce O3, butbecause plume sampling occurred just after emission, suffi-cient time was not allowed for O3 production to occur. Theseobservations for O3 production were confirmed during BOR-TAS from the analysis of ozone photochemistry observedby the FAAM BAe-146 research aircraft (Parrington et al.,2013).

Since the only other major source of tropospheric O3 isfrom anthropogenic pollution (smog), which is primarily re-stricted to the lower troposphere (ground level, boundary la-yar), a simplistic way to demonstrate that boreal biomassburning plumes are producing O3 is to analyse the annualtrend for O3 in the middle to upper troposphere in the North-ern Hemisphere and observe whether increases in O3 VMRvalues coincide with the lofted emissions originating fromfire activity. As indicated in the introduction of this work,ACE-FTS measurements have a bias such that occultationsonly measure down to 5 km above the Earth’s surface, thuslimiting ACE-FTS measurements to the middle to upper tro-posphere. Figure4 plots the monthly median VMRs for O3

Atmos. Chem. Phys., 13, 4529–4541, 2013 www.atmos-chem-phys.net/13/4529/2013/

K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS 4537K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS 9

Fig. 5. HCN VMR profiles from occultations containing measurementsof young (<2 days in age) Canadian boreal biomass burning plumesrecorded during the BORTAS campaign. Each demonstrates theintrusion of pyrogenic emissions into the stratosphere, suggesting thatfire-aided convective processes may cause perturbations that result in the stratosphere-troposphere exchange of air masses.

tion for the set of measurements. As expected, O3 VMRsare at their lowest during the Winter months when biomass650

burning activity is minimal, but elevated values are clearlynoted during the Spring and Summer months. By using theMODIS fire and IASI CO data, the geographic locations forthe sources of the biomass burning can be deduced. The en-hancement of O3 in March is coincident with large pyrogenic655

outflows resulting from slash and burn activity and the burn-ing of crop residues in the southeastern region of mainlandChina, Indochina and India. The median VMR declines inApril and, once again, become elevated in May through toAugust. The burning activity occurring in May is also at-660

tributed to the burning of crop residues, but predominantlyfrom the area in the arable regions along the southern bor-der of Russia. The elevation in O3 in June, July and Augustcorresponds primarily to boreal forest fires. Fig. 4 indicatesthat there is a correlation between fire activity in the Northern665

Hemisphere and O3 production in the free troposphere. Thiswould indicate that the elevated, off-plume, values observedin ss42910 are due to boreal fire activity prior to the time ofmeasurement and that O3 production occurs in the pyrogenicemissions from aged plumes since the net production of O3670

in nascent and young plumes is negative.

5 Stratosphere-Troposphere Exchange Due to Pyrocon-vective Updrafts

Are aged plumes or young plumes which have interactedwith anthropogenic pollution the only source of O3? Dupont675

et al. (2012) made the observation that, in addition to pho-tochemical production of O3 from precursors formed in theplumes, exchange between the stratosphere and the tropo-sphere was a major factor influencing O3 concentrations inplumes measured during ARCTAS. Biomass burning emis-680

sions can be injected in the upper troposphere by isolatedconvection that is not fire-related as well as by pyrocon-vection; are fire-aided convective processes capable of per-turbing the tropopause and permitting O3 from stratosphericair into the free troposphere? During the BORTAS cam-685

paign, ACE-FTS made measurements of boreal plumes overCanada where the pyrogenic outflow had been lofted to alti-tudes where plumes demonstrated intrusion into the strato-sphere. Fig. 5 shows the HCN VMR profiles from sixrandomly selected occultations containing measurements of690

young (<2 days in age) plumes obtained during BORTAS,where enhanced concentrations of HCN are observed at andabove the tropopause. Again, the thermal tropopause heightswere obtained from derived meteorological products (DMPs)that are produced using the ACE-FTS Version 3.0 dataset695

Fig. 5.HCN VMR profiles from occultations containing measurements of young (< 2 days in age) Canadian boreal biomass burning plumesrecorded during the BORTAS campaign. Each demonstrates the intrusion of pyrogenic emissions into the stratosphere, suggesting that fire-aided convective processes may cause perturbations that result in the stratosphere–troposphere exchange of air masses.

in the Northern Hemisphere from all filtered ACE-FTS mea-surements, without the threshold on HCN applied, made inthe troposphere during a 4 yr period (2008–2011). The errorbars represent the calculated 1-sigma standard deviation forthe set of measurements. As expected, O3 VMRs are at theirlowest during the winter months when biomass burning ac-tivity is minimal, but elevated values are clearly noted duringthe spring and summer months. By using the MODIS fire andIASI CO data, the geographic locations for the sources of thebiomass burning can be deduced. The enhancement of O3 inMarch is coincident with large pyrogenic outflows resultingfrom slash and burn activity and the burning of crop residuesin the southeastern region of mainland China, Indochina, andIndia. The median VMR declines in April and, once again,become elevated in May through to August. The burning ac-tivity occurring in May is also attributed to the burning ofcrop residues, but predominantly from the area in the arableregions along the southern border of Russia. The elevationin O3 in June, July, and August corresponds primarily to bo-real forest fires. Figure4 indicates that there is a correlationbetween fire activity in the Northern Hemisphere and O3 pro-duction in the free troposphere. This would indicate that theelevated, off-plume values observed in ss42910 are due toboreal fire activity prior to the time of measurement and thatO3 production occurs in the pyrogenic emissions from aged

plumes since the net production of O3 in nascent and youngplumes is negative.

5 Stratosphere–troposphere exchange due topyroconvective updrafts

Are aged plumes or young plumes which have interactedwith anthropogenic pollution the only source of O3? Dupontet al. (2012) made the observation that, in addition to pho-tochemical production of O3 from precursors formed in theplumes, exchange between the stratosphere and the tropo-sphere was a major factor influencing O3 concentrations inplumes measured during ARCTAS. Biomass burning emis-sions can be injected into the upper troposphere by iso-lated convection that is not fire related as well as by pyro-convection; are fire-aided convective processes capable ofperturbing the tropopause and permitting O3 from strato-spheric air into the free troposphere? During the BORTAScampaign, ACE-FTS made measurements of boreal plumesover Canada, where the pyrogenic outflow had been loftedto altitudes where plumes demonstrated intrusion into thestratosphere. Figure5 shows the HCN VMR profiles fromsix randomly selected occultations containing measurementsof young (< 2 days in age) plumes obtained during BOR-TAS, where enhanced concentrations of HCN are observedat and above the tropopause. Again, the thermal tropopause

www.atmos-chem-phys.net/13/4529/2013/ Atmos. Chem. Phys., 13, 4529–4541, 2013

4538 K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS10 K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS

Fig. 6. Observation of strat-trop mixing in Canadian boreal biomass burning plumes. Comparison of the average of 15 VMR profilesofpyrogenic species (HCN, C2H2) measured in the stratosphere to an equivalent number off-plume measurements made over Canada during theSummer (JJA) of 2008-2011. Also compared, are the on- and off-plume average measurements tropospheric VMRs of stratospheric species(HF, HCl) observed in the troposphere using the same occultations. Occultations containing measurements of biomass burning plumescontain elevated VMRs of HCN and C2H2 in the stratosphere and enhanced HF and HCl in the troposphere, which give an indication of theexchange of air masses between the troposphere and stratosphere likely due to perturbation caused by fire-aided convective processes.

(Manney et al., 2007).An inspection of the HCN VMR profiles of the boreal

biomass burning plumes recorded over Canada during theSummer months (JJA) of 2008-2011 show that a number ofmeasurements containing enhancements in HCN above the700

tropopause were observed. They are more prevalent in occul-tations made in high latitudes where the tropopause is closerto ground level. Injection of pyrogenic emissions into thestratosphere would be facilitated for instances of boreal fireactivity occuring in the northerly latitudes of Siberia, Alaska705

and Northern Canada. Given the evidence for the occur-rence of stratospheric intrusion of biomass burning plumesdue to fire-aided convective processes, the question remainswhether or not these intrusions cause perturbations in whichstratospheric air masses are drawn into the free troposphere,710

potentially mixing stratospheric O3 with the plume emis-sions, which may account for the sporadic nature of the pres-ence of elevated O3 in young boreal biomass burning plumesobserved in prior measurement campaigns.

If perturbations are indeed occurring, there should be ev-715

idence of enhancements in the VMRs of pyrogenic species

originating from these lofted plumes in the stratosphere whencompared to off-plume measurements made over Canadaduring the same time period. Conversely, it is expected thatthere would be elevations in the VMRs of typically strato-720

spheric species in the upper troposphere as stratospheric airmasses are being forced down into the troposphere as a resultof these pyroconvective updrafts. Fig. 6 contains the VMRprofiles for two pyrogenic species (C2H2 and HCN) mea-sured in the stratosphere and for two stratospheric species725

(HF and HCl) in the troposphere. These profiles were ob-tained by averaging the profiles measured from above and be-low the tropopause accordingly for 15 occultations contain-ing measurements of young (<2 days in age) boreal biomassburning plumes recorded over Canada during the Summer730

months (JJA) of 2008-2011, which demonstrated in theirindividual HCN concentration profiles an intrusion of thebiomass burning plume into the stratosphere. These averagedprofiles are then compared to the averaged profiles of thesame number of off-plume measurements taken over Canada735

during the same time period. We clearly observe in Fig. 6an enhancement in the profiles for the pyrogenic species in

Fig. 6. Observation of strat.–trop. mixing in Canadian boreal biomass burning plumes. Comparison of the average of 15 VMR profiles ofpyrogenic species (HCN, C2H2) measured in the stratosphere to an equivalent number off-plume measurements made over Canada during thesummer (JJA) of 2008–2011. Also compared are the on- and off-plume average measurements tropospheric VMRs of stratospheric species(HF, HCl) observed in the troposphere using the same occultations. Occultations containing measurements of biomass burning plumescontain elevated VMRs of HCN and C2H2 in the stratosphere and enhanced HF and HCl in the troposphere, which give an indication of theexchange of air masses between the troposphere and stratosphere likely due to perturbation caused by fire-aided convective processes.

heights were obtained from derived meteorological products(DMPs) that are produced using the ACE-FTS version 3.0dataset (Manney et al., 2007).

An inspection of the HCN VMR profiles of the borealbiomass burning plumes recorded over Canada during thesummer months (JJA) of 2008–2011 show that a numberof measurements containing enhancements in HCN abovethe tropopause were observed. They are more prevalent inoccultations made in high latitudes where the tropopauseis closer to ground level. Injection of pyrogenic emissionsinto the stratosphere would be facilitated for instances ofboreal fire activity occurring in the northerly latitudes ofSiberia, Alaska, and Northern Canada. Given the evidencefor the occurrence of stratospheric intrusion of biomass burn-ing plumes due to fire-aided convective processes, the ques-tion remains whether or not these intrusions cause perturba-tions in which stratospheric air masses are drawn into thefree troposphere, potentially mixing stratospheric O3 withthe plume emissions, which may account for the sporadic na-ture of the presence of elevated O3 in young boreal biomassburning plumes observed in prior measurement campaigns.

If perturbations are indeed occurring, there should be ev-idence of enhancements in the VMRs of pyrogenic speciesoriginating from these lofted plumes in the stratosphere whencompared to off-plume measurements made over Canadaduring the same time period. Conversely, it is expected thatthere would be elevations in the VMRs of typically strato-spheric species in the upper troposphere as stratospheric airmasses are being forced down into the troposphere as a resultof these pyroconvective updrafts. Figure6 contains the VMRprofiles for two pyrogenic species (C2H2 and HCN) mea-sured in the stratosphere and for two stratospheric species(HF and HCl) in the troposphere. These profiles were ob-tained by averaging the profiles measured from above and be-low the tropopause accordingly for 15 occultations contain-ing measurements of young (< 2 days in age) boreal biomassburning plumes recorded over Canada during the summermonths (JJA) of 2008–2011, which demonstrated in theirindividual HCN concentration profiles an intrusion of thebiomass burning plume into the stratosphere. These averagedprofiles are then compared to the averaged profiles of thesame number of off-plume measurements taken over Canadaduring the same time period. We clearly observe in Fig.6 an

Atmos. Chem. Phys., 13, 4529–4541, 2013 www.atmos-chem-phys.net/13/4529/2013/

K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS 4539K.A. Tereszchuket al.: ACE-FTS Observations of Pyrogenic Trace Species During BORTAS 11

Fig. 7. Observation of stratospheric O3 in the free troposphere.Comparison of the averaged O3 VMR profiles recorded from be-low the tropopause for the six young boreal plumes in Fig. 5, whichdemonstrated intrusion of pyrogenic emissions into the strato-sphere, to the averaged profiles of six plumes where their HCNprofiles indicate that the pyrogenic outflow was confined to the tro-posphere. The VMR profile obtained from the plumes recorded inFig. 5 clearly indicates a large increases in O3 relative to the pro-file where intrusion has not occurred. This strongly suggests thatstratospheric air is being forced into the free troposphereand thatthe origin of the O3 in these young plumes is from the stratosphere.

the stratosphere when compared to the averaged off-plumemeasurements, and similarly, enhancements are observed inthe concentrations of HF and HCl in the plume measure-740

ments made within the troposphere, indicating that pyrocon-vection and/or fire-aided convective processes are causinganexchange in air masses between the troposphere and strato-sphere. To demonstrate that O3 from the stratosphere is in-deed being injected into the troposphere, the O3 profiles from745

data recorded below the tropopause from the six occultationsreported in Fig. 5 were averaged together and compared to ameasurements of biomass burning plumes of the same rela-tive age from a similar number of occultations recorded dur-ing the BORTAS campaign, where the HCN profiles show750

maximum values in the middle troposphere and do not showevidence of intrusion into the stratosphere. Upon compari-son, we observe in Fig. 7 that the enhancements in the VMRvalues of O3 are quite evident in the profile where strato-spheric intrusion of the biomass burning plume is occurring,755

strongly suggesting that stratosphere-troposphere exchangeis taking place and injecting stratospheric air through thetropopause and mixing with the pyrogenic outflows fromthese boreal fires.

6 Conclusions760

Measurements of trace species from biomass burning plumeswere made using (ACE-FTS). An extensive set of 14molecules, CH3OH, C2H2, C2H6, C3H6O, CO, HCN,HCOOH, HNO3, H2CO, NO, NO2, OCS, O3 and PAN havebeen analyzed to generate sets of age-dependent enhance-765

ment ratios calculated from the pyrogenic emissions fromboreal fires originating from both North America (Canada,Alaska) and Siberia for a period of up to seven days. Theseenhancement ratios can be used to test the predictions ofchemical transport models used to study the evolution of py-770

rogenic species in the long-range transport of boreal biomassburning plumes in the middle and upper troposphere. Theenhancement ratios derived for the pyrogenic species mea-sured in the emissions from the boreal fires exhibit the ex-pected trend where the ratio values for the shorter-lived pri-775

mary pyrogenic species decrease over time as these speciesare oxidized primarily by the OH radical, as the plumeages and values for longer-lived species such as HCN andC2H6 remain relatively unchanged. Increasing negative val-ues are observed for the oxidant species, including O3, in-780

dicating that these species are undergoing a destruction pro-cess in the plume as it ages such that concentrations of theoxidant species have dropped below their off-plume val-ues. Even though decreases in O3 concentrations are ob-served within the plumes, the off-plume concentrations of785

O3 measured during the height of boreal forest fire activ-ity in Canada are elevated above typical background val-ues. Median concentrations of O3 in the middle and up-per troposphere of the Northern Hemisphere demonstrate astrong seasonality where enhancements in O3 are coincident790

with biomass burning activity, indicating that biomass burn-ing is generating O3, but as a secondary pyrogenic species inaged plumes. Results from previous measurement campaignshave indicated that elevated concentrations of O3 and posi-tive molar ratios of∆O3/∆CO have been observed in young795

boreal biomass burning plumes, but the presence of O3 in theyoung plumes has been attributed to pollution effects wherethe pyrogenic outflows have mixed with either local urbanNOx emissions or pyrogenic emissions from the long-rangetransport of older plumes. The results from this work have800

demonstrated that an additional source of O3 measured inyoung boreal plumes may be attributed to the occurrence ofstratosphere-troposphere exchange due to the pyroconvectiveupdrafts from large fires. Perturbations caused by the loftedemissions in these fire-aided convective processes can poten-805

tially result in the intrusion of stratospheric air masses intothe free troposphere and subsequent mixing of stratosphericO3 into these pyrogenic outflows.

Acknowledgements. The ACE mission is funded primarily by theCanadian Space Agency. The authors would like to thank the Natu-810

ral Environment Research Council of the United Kingdom for fund-ing K.A. Tereszchuk through Grant NE/F017391/1 and the WildFund for their support of G. Gonzalez Abad. The authors would

Fig. 7. Observation of stratospheric O3 in the free troposphere.Comparison of the averaged O3 VMR profiles recorded from be-low the tropopause for the six young boreal plumes in Fig.5, whichdemonstrated intrusion of pyrogenic emissions into the strato-sphere, to the averaged profiles of six plumes where their HCNprofiles indicate that the pyrogenic outflow was confined to the tro-posphere. The VMR profile obtained from the plumes recorded inFig. 5 clearly indicates large increases in O3 relative to the profilewhere intrusion has not occurred. This strongly suggests that strato-spheric air is being forced into the free troposphere and that theorigin of the O3 in these young plumes is from the stratosphere.

enhancement in the profiles for the pyrogenic species in thestratosphere when compared to the averaged off-plume mea-surements and, similarly, enhancements are observed in theconcentrations of HF and HCl in the plume measurementsmade within the troposphere, indicating that pyroconvectionand/or fire-aided convective processes are causing an ex-change in air masses between the troposphere and strato-sphere. To demonstrate that O3 from the stratosphere is in-deed being injected into the troposphere, the O3 profiles fromdata recorded below the tropopause from the six occultationsreported in Fig.5 were averaged together and compared tomeasurements of biomass burning plumes of the same rela-tive age from a similar number of occultations recorded dur-ing the BORTAS campaign, where the HCN profiles showmaximum values in the middle troposphere and do not showevidence of intrusion into the stratosphere. Upon comparisonwe observe in Fig.7 that the enhancements in the VMR val-ues of O3 are quite evident in the profile where stratosphericintrusion of the biomass burning plume is occurring, stronglysuggesting that stratosphere–troposphere exchange is takingplace and injecting stratospheric air through the tropopauseand mixing with the pyrogenic outflows from these borealfires.

6 Conclusions

Measurements of trace species from biomass burning plumeswere made using ACE-FTS. An extensive set of 14molecules – CH3OH, C2H2, C2H6, C3H6O, CO, HCN,HCOOH, HNO3, H2CO, NO, NO2, OCS, O3, and PAN– have been analysed to generate sets of age-dependentenhancement ratios calculated from the pyrogenic emis-sions from boreal fires originating from both North America(Canada, Alaska) and Siberia for a period of up to 7 days.These enhancement ratios can be used to test the predictionsof CTMs used to study the evolution of pyrogenic species inthe long-range transport of boreal biomass burning plumes inthe middle and upper troposphere. The enhancement ratiosderived for the pyrogenic species measured in the emissionsfrom the boreal fires exhibit the expected trend where theratio values for the shorter lived primary pyrogenic speciesdecrease over time as these species are oxidized primarilyby the OH radical, as the plume ages and values for longerlived species such as HCN and C2H6 remain relatively un-changed. Increasing negative values are observed for the ox-idant species, including O3, indicating that these species areundergoing a destruction process in the plume as it ages suchthat concentrations of the oxidant species have dropped be-low their off-plume values. Even though decreases in O3 con-centrations are observed within the plumes, the off-plumeconcentrations of O3 measured during the height of bo-real forest fire activity in Canada are elevated above typi-cal background values. Median concentrations of O3 in themiddle and upper troposphere of the Northern Hemispheredemonstrate a strong seasonality where enhancements in O3are coincident with biomass burning activity, indicating thatbiomass burning is generating O3, but as a secondary pyro-genic species in aged plumes. Results from previous mea-surement campaigns have indicated that elevated concentra-tions of O3 and positive molar ratios of1O3/1CO have beenobserved in young boreal biomass burning plumes, but thepresence of O3 in the young plumes has been attributed topollution effects where the pyrogenic outflows have mixedwith either local urban NOx emissions or pyrogenic emis-sions from the long-range transport of older plumes. The re-sults from this work have demonstrated that an additionalsource of O3 measured in young boreal plumes may be at-tributed to the occurrence of stratosphere–troposphere ex-change due to the pyroconvective updrafts from large fires.Perturbations caused by the lofted emissions in these fire-aided convective processes can potentially result in the intru-sion of stratospheric air masses into the free troposphere andsubsequent mixing of stratospheric O3 into these pyrogenicoutflows.

www.atmos-chem-phys.net/13/4529/2013/ Atmos. Chem. Phys., 13, 4529–4541, 2013

4540 K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS

Acknowledgements.The ACE mission is funded primarily bythe Canadian Space Agency. The authors would like to thank theNatural Environment Research Council of the United Kingdomfor funding K. A. Tereszchuk through grant NE/F017391/1 andthe Wild Fund for their support of G. Gonzalez Abad. The authorswould like to further acknowledge the LATMOS research groupfor providing the Level 2 IASI CO data. IASI has been developedand built under the responsibility of the Centre National d’EtudesSpatiales (CNES, France). It is flown onboard the MetOp satellitesas part of the Eumetsat Polar System.

Edited by: S. Matthiesen

References

Akagi, S. K., Yokelson, R. J., Wiedinmyer, C., Alvarado, M. J.,Reid, J. S., Karl, T., Crounse, J. D., and Wennberg, P. O.: Emis-sion factors for open and domestic biomass burning for usein atmospheric models, Atmos. Chem. Phys., 11, 4039–4072,doi:10.5194/acp-11-4039-2011, 2011.

Allen, G., Remedios, J. J., Newnham, D. A., Smith, K. M., andMonks, P. S.: Improved mid-infrared cross-sections for perox-yacetyl nitrate (PAN) vapour, Atmos. Chem. Phys., 5, 47–56,doi:10.5194/acp-5-47-2005, 2005.

Alvarado, M. J., Logan, J. A., Mao, J., Apel, E., Riemer, D., Blake,D., Cohen, R. C., Min, K.-E., Perring, A. E., Browne, E. C.,Wooldridge, P. J., Diskin, G. S., Sachse, G. W., Fuelberg, H.,Sessions, W. R., Harrigan, D. L., Huey, G., Liao, J., Case-Hanks,A., Jimenez, J. L., Cubison, M. J., Vay, S. A., Weinheimer, A.J., Knapp, D. J., Montzka, D. D., Flocke, F. M., Pollack, I.B., Wennberg, P. O., Kurten, A., Crounse, J., Clair, J. M. St.,Wisthaler, A., Mikoviny, T., Yantosca, R. M., Carouge, C. C.,and Le Sager, P.: Nitrogen oxides and PAN in plumes from borealfires during ARCTAS-B and their impact on ozone: an integratedanalysis of aircraft and satellite observations, Atmos. Chem.Phys., 10, 9739–9760,doi:10.5194/acp-10-9739-2010, 2010.

Andreae, M. O. and Merlet, P.: Emission of trace gases and aerosolsfrom biomass burning, G. Biogeochem. Cy., 15, 955–966, 2001.

Bernath, P. F., McElroy, C. T., Abrams, M. C., Boone, C. D., Butler,M., Camy-Peyret, C., Carleer, M., Clerbaux, C., Coheur, P.-F.,Colin, R., DeCola, P., DeMaziere, M., Drummond, J. R., Du-four, D., Evans, W. F. J., Fast, H., Fussen, D., Gilbert, K., Jen-nings, D. E., Llewellyn, E. J., Lowe, R. P., Mahieu, E., Mc-Connell, J. C., McHugh, M., McLeod, S. D., Michaud, R., Mid-winter, C., Nassar, R., Nichitiu, F., Nowlan, C., Rinsland, C. P.,Rochon, Y. J., Rowlands, N., Semeniuk, K., Simon, P., Skel-ton, R., Sloan, J. J., Soucy, M.-A., Strong, K., Tremblay, P.,Turnbull, D., Walker, K.A., Walkty, I., Wardle, D. A., Wehrle,V., Zander, R., and Zou, J.: Atmospheric Chemistry Experiment(ACE): Mission overview, Geophys. Res. Lett., 32, L15S01,doi:10.1029/2005GL022386, 2005.

Boone, C. D., Nassar, R., Walker, K. A., Rochon, Y., McLeod, S. D.,Rinsland, C. P., and Bernath, P. F.: Retrievals for the atmosphericchemistry experiment Fourier-transform spectrometer, Appl. Op-tics, 44, 7218–7231, 2005.

Clerbaux, C., Boynard, A., Clarisse, L., George, M., Hadji-Lazaro,J., Herbin, H., Hurtmans, D., Pommier, M., Razavi, A., Turquety,S., Wespes, C., and Coheur, P.-F.: Monitoring of atmosphericcomposition using the thermal infrared IASI/MetOp sounder, At-

mos. Chem. Phys., 9, 6041–6054,doi:10.5194/acp-9-6041-2009,2009.

Coheur, P.-F., Herbin, H., Clerbaux, C., Hurtmans, D., Wespes, C.,Carleer, M., Turquety, S., Rinsland, C. P., Remedios, J., Hauglus-taine, D., Boone, C. D., and Bernath, P. F.: ACE-FTS observationof a young biomass burning plume: first reported measurementsof C2H4, C3H6O, H2CO and PAN by infrared occultation fromspace, Atmos. Chem. Phys., 7, 5437–5446,doi:10.5194/acp-7-5437-2007, 2007.

Coheur, P.-F., Clarisse, L., Turquety, S., Hurtmans, D., and Cler-baux, C.: IASI measurements of reactive trace species inbiomass burning plumes, Atmos. Chem. Phys., 9, 5655–5667,doi:10.5194/acp-9-5655-2009, 2009.

Crutzen, P. J., Height, L. E., Krasnec, J. P., Pollock, W. P., and Seiler,W.: Biomass burning as a source of atmospheric gases CO, H2,N2O, NO, CH3Cl, and COS, Nature, 282, 253–256, 1979.

Crutzen, P. J. and Andreae, M. O.: Biomass burning in the trop-ics: Impact on atmospheric chemistry and biogeochemical cy-cles, Science, 250, 1669–1678, 1990.

Draxler, R. R., and Rolph, G. D.: HYSPLIT (HYbrid Single-ParticleLagrangian Integrated Trajectory) Model, NOAA Air ResourcesLaboratory, Silver Spring, MD, available at:http://www.arl.noaa.gov/ready/hysplit4.html(last access: October 2012), 2003.

Dupont, R., Pierce, B., Worden, J., Hair, J., Fenn, M., Hamer,P., Natarajan, M., Schaack, T., Lenzen, A., Apel, E., Dibb, J.,Diskin, G., Huey, G., Weinheimer, A., Kondo, Y., and Knapp,D.: Attribution and evolution of ozone from Asian wild fires us-ing satellite and aircraft measurements during the ARCTAS cam-paign, Atmos. Chem. Phys., 12, 169–188,doi:10.5194/acp-12-169-2012, 2012.

Fiedler, V., Arnold, F., Ludmann, S., Minikin, A., Hamburger,T., Pirjola, L., Dornbrack, A., and Schlager, H.: Africanbiomass burning plumes over the Atlantic: aircraft based mea-surements and implications for H2SO4 and HNO3 mediatedsmoke particle activation, Atmos. Chem. Phys., 11, 3211–3225,doi:10.5194/acp-11-3211-2011, 2011.

George, M., Clerbaux, C., Hurtmans, D., Turquety, S., Coheur, P.-F., Pommier, M., Hadji-Lazaro, J., Edwards, D. P., Worden, H.,Luo, M., Rinsland, C., and McMillan, W.: Carbon monoxide dis-tributions from the IASI/METOP mission: evaluation with otherspace-borne remote sensors, Atmos. Chem. Phys., 9, 8317–8330,doi:10.5194/acp-9-8317-2009, 2009.

Glatthor, N., von Clarmann, T., Stiller, G. P., Funke, B., Kouk-ouli, M. E., Fischer, H., Grabowski, U., Hopfner, M., Kellmann,S., and Linden, A.: Large-scale upper tropospheric pollutionobserved by MIPAS HCN and C2H6 global distributions, At-mos. Chem. Phys., 9, 9619–9634,doi:10.5194/acp-9-9619-2009,2009.

Hobbs, P. V., Sinha, P., Yokelson, R. J., Christian, T. J., Blake, D. R.,Gao, S., Kirchstetter, T. W., Novakov, T., and Pilewskie, P.: Evo-lution of gases and particles from a savanna fire in South Africa,J. Geophys. Res., 108, 8485,doi:10.1029/2002JD002352, 2003.

Hornbrook, R. S., Blake, D. R., Diskin, G. S., Fried, A., Fuelberg,H. E., Meinardi, S., Mikoviny, T., Richter, D., Sachse, G. W.,Vay, S. A., Walega, J., Weibring, P., Weinheimer, A. J., Wiedin-myer, C., Wisthaler, A., Hills, A., Riemer, D. D., and Apel, E. C.:Observations of nonmethane organic compounds during ARC-TAS – Part 1: Biomass burning emissions and plume enhance-ments, Atmos. Chem. Phys., 11, 11103–11130,doi:10.5194/acp-

Atmos. Chem. Phys., 13, 4529–4541, 2013 www.atmos-chem-phys.net/13/4529/2013/

K. A. Tereszchuk et al.: ACE-FTS observations of pyrogenic trace species during BORTAS 4541

11-11103-2011, 2011.Jaegle, L., Jacob, D. J., Brune, W. H., Faloona, I. C., Tan, D., Kondo,

Y., Sachse, G. W., Anderson, B., Gregory, G. L., Vay, S., Singh,H. B., Blake, D. R., and Shetter, R.: Ozone production in theupper troposphere and the influence of aircraft during SONEX:Approach of NOx-saturated conditions, Geophys. Res. Lett., 26,3081–3084, 1999.

Jaffe, D. A. and Wigder, N. L.: Ozone production from wildfires: ACritical Review, Atmos. Environ., 51, 1–10, 2012.

Lefer, B. L., Talbot R. W., Harriss, R. H., Bradshaw, J. D., Sand-holm, S. T., Olson, J. O., Sachse, G.W., Collins, J., Shipham, M.A., Blake, D. R., Klemm, K. I., Klemm, O., Gorzelska, K., andBarrick, J.: Enhancement of acidic gases in biomass burning im-pacted air masses over Canada, J. Geophys. Res., 99, 1721–1737,1994.

Li, Q., Jacob, D. J., Yantosca, R. M., Heald, C. L., Singh,H. B., Koike, M., Zhao, Y., Sachse, G. W., and Streets, D.G.: A global three-dimensional model analysis of the atmo-spheric budgets of HCN and CH3CN: Constraints from air-craft and ground measurements, J. Geophys. Res., 108, 8827,doi:10.1029/2002JD003075, 2003.

Manney, G. L., Daffer, W. H., Zawodny, J. M., Bernath, P. F., Hop-pel, K. W.,Walker, K. A., Knosp, B. W., Boone, C., Remsberg,E. E., Santee, M. L., Harvey, V. L., Pawson, S., Jackson, D. R.,Deaver, L., McElroy, C. T., McLinden, C. A., Drummond, J. R.,Pumphrey, H. C., Lambert, A., Schwartz, M. J., Froidevaux, L.,McLeod, S., Takacs, L. L., Suarez, M. J., Trepte, C. R., Cuddy,D. C., Livesey, N. J., Harwood, R. S., and Waters, J. W. : Solaroccultation satellite data and derived meteorological products:sampling issues and comparisons with Aura MLS, J. Geophys.Res., 112, D24S50,doi:10.1029/2007JD008709, 2007.

Palmer, P. I., Parrington, M., Lee, J. D., Lewis, A. C., Rickard, A.R., Bernath, P. F., Duck, T. J., Waugh, D. L., Tarasick, D. W., An-drews, S., Aruffo, E., Bailey, L. J., Barrett, E., Bauguitte, S. J.-B., Curry, K. R., Di Carlo, P., Chisholm, L., Dan, L., Forster, G.,Franklin, J. E., Gibson, M. D., Griffin, D., Helmig, D., Hopkins,J. R., Hopper, J. T., Jenkin, M. E., Kindred, D., Kliever, J., LeBreton, M., Matthiesen, S., Maurice, M., Moller, S., Moore, D.P., Oram, D. E., O’Shea, S. J., Christopher Owen, R., Pagniello,C. M. L. S., Pawson, S., Percival, C. J., Pierce, J. R., Punjabi,S., Purvis, R. M., Remedios, J. J., Rotermund, K. M., Sakamoto,K. M., da Silva, A. M., Strawbridge, K. B., Strong, K., Taylor,J., Trigwell, R., Tereszchuk, K. A., Walker, K. A., Weaver, D.,Whaley, C., and Young, J. C.: Quantifying the impact of BORealforest fires on Tropospheric oxidants over the Atlantic using Air-craft and Satellites (BORTAS) experiment: design, execution andscience overview, Atmos. Chem. Phys. Discuss., 13, 4127–4181,doi:10.5194/acpd-13-4127-2013, 2013.

Parrington, M., Palmer, P. I., Lewis, A. C., Lee, J. D., Rickard, A.R., Di Carlo, P., Taylor, J. W., Hopkins, J. R., Punjabi, S., Oram,D. E., Forster, G., Aruffo, E., Moller, S. J., Bauguitte, S. J.-B.,Allan, J. D., Coe, H., and Leigh, R. J.: Ozone photochemistry inboreal biomass burning plumes, Atmos. Chem. Phys. Discuss.,13, 1795–1853,doi:10.5194/acpd-13-1795-2013, 2013.

Rinsland, C. P., Dufour, G., Boone, C. D., Bernath, P. F., and Chiou,L.: Atmospheric Chemistry Experiment (ACE) measurements ofelevated Southern Hemisphere upper tropospheric CO, C2H6,HCN, and C2H2 mixing ratios from biomass burning emis-sions and long-range transport, Geophys. Res. Lett., 32, L20803,

doi:10.1029/2005GL024214, 2005.Rinsland, C. P., Dufour, G., Boone, C. D., Bernath, P. F., Chiou,

L., Coheur, P. -F., Turquety, S., and Clerbaux, C.: Satellite bo-real measurements over Alaska and Canada during June–July2004: Simultaneous measurements of upper tropospheric CO,C2H6, HCN, CH3Cl, CH4, C2H2, CH3OH, HCOOH, OCS,and SF6 mixing ratios, Global Biogeochem. Cy., 21, GB3008,doi:10.1029/2006GB002795, 2007.

Simpson, I. J., Akagi, S. K., Barletta, B., Blake, N. J., Choi, Y.,Diskin, G. S., Fried, A., Fuelberg, H. E., Meinardi, S., Rowland,F. S., Vay, S. A., Weinheimer, A. J., Wennberg, P. O., Wiebring,P., Wisthaler, A., Yang, M., Yokelson, R. J., and Blake, D. R.:Boreal forest fire emissions in fresh Canadian smoke plumes:C1-C10 volatile organic compounds (VOCs), CO2, CO, NO2,NO, HCN and CH3CN, Atmos. Chem. Phys., 11, 6445–6463,doi:10.5194/acp-11-6445-2011, 2011.

Singh, H. B., Salas, L., Herlth, D., Kolyer, R., Czech, E., Viezee,W., Li, Q., Jacob, D. J., Blake, D., Sachse, G., Harward,C. N., Fuelberg, H., Kiley, C. M., Zhao, Y., and Kondo, Y.:In situ measurements of HCN and CH3CN over the PacificOcean: Sources, sinks, and budgets, J. Geophys. Res., 108, 8795,doi:10.1029/2002JD003006, 2003.

Singh, H. B., Anderson, B. E., Brune, W. H., Cai, C., Cohen, R. C.,Crawford, J. H., Cubison, M. J., Czech, E. P., Emmons, L., Fuel-berg, H. E., Huey, G., Jacob, D. J, Jimenez, J. L., Kaduwela, A.,Kondo, Y., Mao, J., Olson, J. R., Sachse, G. W., and Vay, S. A.,Weinheimer, A., Wennberg, P. O., Wisthalerm, A., the ARCTASScience Team: Pollution influences on atmospheric compositionand chemistry at high northern latitudes: Boreal and Californiaforest fire emissions, Atmos. Environ., 44, 4553–4564, 2010.

Stubenrauch, C. J., Cros, S., Guignard, A., and Lamquin, N.:A 6-year global cloud climatology from the Atmospheric In-fraRed Sounder AIRS and a statistical analysis in synergy withCALIPSO and CloudSat, Atmos. Chem. Phys., 10, 7197–7214,doi:10.5194/acp-10-7197-2010, 2010.

Torres, O., Chen, Z., Jethva, H., Ahn, C., Freitas, S. R., and Bhartia,P. K.: OMI and MODIS observations of the anomalous 2008–2009 Southern Hemisphere biomass burning seasons, Atmos.Chem. Phys., 10, 3505–3513,doi:10.5194/acp-10-3505-2010,2010.

Tereszchuk, K. A., Gonzalez Abad, G., Clerbaux, C., Hurtmans,D., Coheur, P.-F., and Bernath, P. F.: ACE-FTS measurements oftrace species in the characterization of biomass burning plumes,Atmos. Chem. Phys., 11, 12169–12179,doi:10.5194/acp-11-12169-2011, 2011.

Turquety, S., Hurtmans, D., Hadji-Lazaro, J., Coheur, P.-F., Cler-baux, C., Josset, D., and Tsamalis, C.: Tracking the emissionand transport of pollution from wildfires using the IASI CO re-trievals: analysis of the summer 2007 Greek fires, Atmos. Chem.Phys., 9, 4897–4913,doi:10.5194/acp-9-4897-2009, 2009.

Yokelson, R. J., Crounse, J. D., DeCarlo, P. F., Karl, T., Urbanski,S., Atlas, E., Campos, T., Shinozuka, Y., Kapustin, V., Clarke,A. D., Weinheimer, A., Knapp, D. J., Montzka, D. D., Holloway,J., Weibring, P., Flocke, F., Zheng, W., Toohey, D., Wennberg, P.O., Wiedinmyer, C., Mauldin, L., Fried, A., Richter, D., Walega,J., Jimenez, J. L., Adachi, K., Buseck, P. R., Hall, S. R., andShetter, R.: Emissions from biomass burning in the Yucatan, At-mos. Chem. Phys., 9, 5785–5812,doi:10.5194/acp-9-5785-2009,2009.

www.atmos-chem-phys.net/13/4529/2013/ Atmos. Chem. Phys., 13, 4529–4541, 2013